DAC8775: STM32F446RE show no communication with DAC8775

Mohammed,

I always recommend looking at the SPI communication with an oscilloscope or a logic analyzer. It's important to see the digital communication between the microcontroller and the device. For this device, I'd look at the SCLK, SDIN, SDO, and /SYNC lines just to see that the communication is exactly what you think it is. Once you can grab a shot of that, post it back to this thread.

Another useful test would be to write to and read back from one of the registers. You could just read from the status register, which should give you a 1000h readback, but I'd also try writing to and reading back from the DAC data register from one of the DACs. Use something other than all 1s or all 0s, you want to make sure you're reading real data.

I didn't see anything wrong with your sequence when I took a quick look at it. This is how I would generally set the sequence:

Reset
Enable buck-boost
Configure buck-boost (or just use default)
Enable reference
Configure DAC (range and maybe code)
Enabel DAC output

If you can, I would consider getting the DAC8775EVM. It has the device and is controlled though USB using a SM-USB-DIG controller. The software GUI is easy to use and you can look at the digital lines going in and out of the device to verify operation against your board and code.

Regardless, start with looking at the SPI communications going between your microcontroller and device on the board. Post back with what you see.

Joseph Wu

Hi Joseph,

Thanks for your suggestion. Today I get the new logic analyzer for SPI analysis. After capturing data it seems all is ok, but there is an error during the reset of the microntroller. 

Error: The Initial state of the clock line dose does not match the setting (It always comes after reset)

Attached Screenshot of the trend from the logic analyzer.

Please advise what's wrong in this..

Mohammad,

Looking at your plots, I don't see anything in particular that is incorrect. I think this is the sequence that you're sending:

010001h DAC reset
060001h Select Buck-boost Converter A selected
070001h Config buck-boost positive enabled, negative disabled
030030h Select DAC 0030 Enable SDO, Channel A selected
04D01Ch Config DAC D01C Current limit 24mA, HART disabled, output enabled, Slew rate enabled, 4-20mA range
05FFFFh Set DAC data register

Additionally, it looks like the SPI is set up so that the data is clocked in on the falling edge of SCLK. Generally, I use the SPI with the SCLK idling low, but that's not a strict requirement based on the timing diagram in Figure 1 of the datasheet. Also, the SPI clock is rather slow, running at about 250kHz, but that shouldn't be a problem either.

I would definitely check the physical connections and make sure the SPI signals are reaching the device. I would also alter your sequence a little bit. Once you enable the SDO with the write to 03h, I would try to read back from the device. Right now, there's no movement on the SDO line, and it would be best to first ensure that there's any communication getting to the device. If your SCLK does not reach the device, it would also cause nothing to happen on SDO. Again make sure that the SDO is enabled after the write to 03h.

Joseph Wu

Hi Joseph,

Thanks for your response.

Yesterday I forget to tell you with above code DAC was responding with  0.32mA & 9.99 Vdc at channel A.

As per request I have capture SDO pin data, I kept the prevous code same just added read function as below

Logic Analyzer Trend (Complete trend for Code-withoud repeating)

Below are the screenshot for each spike after reset.

At 0xFFFF (05h), DAC channel A out- 0.32mA, 9.99 Vdc

At 0x0FFF (05h), DAC Channel A out, 0.03mA, 1.01 Vdc. (This is not included in this logic analyzer data)

Don't know why Voltage is changing although we set it to 4-20mA out.

Don't know why current is so low..

Please advise..

Thanks

Joseph,

There was a mistake on my side in not enabling internal reference voltage.

I set 02h to 0x0013, it starts working....

Thanks once again..